What controls Turbo Core in Xeons?

[CANONKONG]

I meant to quote you originally, you only have a modded BIOS for the AsRock Extreme 4/3.1 not for the original Extreme 4. I was going to try to flash it but in the AsRock website the BIOS are different with different release dates. Which leads to believe they are not the same.

Do you have the original bios about Extreme 4? I don't found this mainboard on the official website,I can help to modify it.

 

[GTZ]

Do you have the original bios about Extreme 4? I don't found this mainboard on the official website,I can help to modify it.

Here is the link to all the BIOS from the official website. Again I really appreciate the and thanks again!!!!

http://www.asrock.com/mb/Intel/X99 Extreme4/?cat=Download&os=BIOS

 

[CANONKONG]

Here is the link to all the BIOS from the official website. Again I really appreciate the and thanks again!!!!

http://www.asrock.com/mb/Intel/X99 Extreme4/?cat=Download&os=BIOS

I had uploaded,you can try it now.
http://pan.baidu.com/s/1dFkuQel

 

[Dufus]

I am a bit surprised your CPU consumes so much power. Got my 2686v3 (same TDP) on 160W in AVX2 with stable x31 Multi.

It's not so bad, Linpack is a hard bench for AVX2. If someone wants to try it using the same version then https://www.mediafire.com/?63mo13dkm6tm0gm

Other links can be taken from here https://forums.tweaktown.com/gigaby...system-info-benchmarking-stability-tools.html

Best results are had if threads are set to number of physical cores and affinity set to one thread per core. Use problem size 30000 for comparison but make sure the heat and power can be handled first.

Are you experiencing an vcore throttle ?

What is that exactly? VCore appears to operate as expected other than not being able to set override mode (Fixed voltage).

However, do you have any experience with 601 ? Maybe it might help us modifying EDP readings.

It's responsible for current limiting but you should be able to set this in the BIOS usually. The Haswell desktop and mobile had a problem with this setting and EDP unless it was set to 256A or higher which basically disabled current limiting. For Xeon you can similarly I guess, mine is set at 1023. EDP however can cover a lot, not just current throttling. For instance if your VID to multi setting is set above a predetermined value when AVX2 is running. For instance I can have EDP and multi throttling when consuming less than 70W with AVX2 or even faking it to half that value.

Im not much interested in boosting DRAM, but of course is a side effect of setting 61E. Which soutbridge registered need to be setted ? Can you specify ?

It's been over 2 years since I looked at that briefly so I would have to go through it again. My Xeon however doesn't have that MSR so hard for me to be enthusiastic about it. The Southbridge side might have been ME, I'm not 100% though.

What register you wrote to increase nominal TDP to 100. Trying to write TDP nominal, limit 1 or 2 always resulted in getting ignored.

Did you read this posted earlier?
http://forum.notebookreview.com/threads/tdp-and-power-limiting-haswell.766743/
While it is based on desktop / mobile Haswell the principle is the same.

Yes MSR read-only responds to mine multiplier would be unlocked, too. However no change took effect yet. Did you try ?

Sorry I don't understand the question, would you elaborate please.

Im quite new to writing MSR register, understand the concept and can reproduce. But I clearly lack the documentation, especially as Intel keeps the most interesting part a secret. Especially about the 6xx-register. Im curious from where you took all the information and knowledge ? Im mostly guessing and experimenting based on the published technical sheets of Intel and experience of others.

Well I'm new to Xeon and have never worked in the computer or software industry so not the best person to ask. Of course there's the SDM, while very informative they can have mistakes. There's the CPU data sheets as well and you may find data omitted that was previously disclosed. For instance you may find the answer to your power limiting problem in the lga-2011 datasheet but not in the 2011-3. It's just a hunch. Other areas to look are projects like coreboot. And of course there is always trial and error. For instance, trying to see what happens if a microcode is removed. So far I haven't had anything go too bad that I could not recover from.

 

[pututu]

Dual CPU EFI should work single CPU as well.

Thanks. If I understand correctly, the file v3x2_payne_120_20.efi should also work on my single xeon motherboard?

 

[Kalistoval]

ASUS &Gigabyte don't need any efi file and ffs file,because then can set up on the bios and work very well. But ASrock & MSI don't have any effective offset about core voltage & cache voltage on the bios,so they need ffs files to mod the bios.If you don't know which ffs you need, you can choose the V3_C-Payne_70_50.ffs,it tested for more than 30 E5v3 CPUs and worked very stable.

Thats very true, I'm running an Msi x99a gaming pro carbon. If I try to manually adjust my offset voltages in my modded bios my 2697 v3 will just sit at 2.6Ghz with no turbo boost. The V3_payne_70_50.efi has boosted all the way up to a consistent x35 and bursts up to x37. I am way over my head in trying to make an efi file let alone inject it into my bios could you make me a modded bios with the ffs injected into it please. I was able to load up the V3_payne_90_50.efi how ever it results in the same turbo boost as the 70_50. Upon experimenting if I max out my DigitAll Power/ basically Load line calibration and disable CStates I am also able to set my Bclk to 105. Between the 70_50, 80_50, and 90_50 the 70_50 has worked the best. I lack the skill to inject that into my bios though.

 

[C_Payne]

I lack the skill to inject that into my bios though

I will prepare some custom bioses in the Next days. Let me know which board exactly and which efi.

 

[C_Payne]

On another Note: does anyone have a code snippet to load uCode or is familiar with the procedure? I would like to add uCode to my drivers to improve OS and Boot compatibility.

 

[CANONKONG]

On another Note: does anyone have a code snippet to load uCode or is familiar with the procedure? I would like to add uCode to my drivers to improve OS and Boot compatibility.

I do not recommend adding microcode into the drivers,because they can not work if they convert to ffs files. Microcode can load by the systems，if your systems lack of the Microcode,you can use tools to install it.
In fact，make more EFI files are needing，like CPU voltage offset -65mv -75mv -85mv.

 

[Cata40]

All ASrock X99 MOD BIOS were made by myself and tested,I think your mainboad can also work well with this bios.

Hello. It looks like it works well, but it still has moments when it hangs something ... we see how it works

 

[Zladimir]

It's not so bad, Linpack is a hard bench for AVX2. If someone wants to try it using the same version then https://www.mediafire.com/?63mo13dkm6tm0gm

https://www.mediafire.com/?63mo13dkm6tm0gm

I tried, you were right. I hit the 180W. However stragenly hot production was a lot less. In average my CPU kept 8°C cooler than stresstest by Prime95 (which hits only 160W).
However heat waste is fully absorbed by my oil bath cooling

What is that exactly? VCore appears to operate as expected other than not being able to set override mode (Fixed voltage).

From what I recall in electrical sheets Haswell has an per-core vcore peak throttle. It allows for ns-spikes but throttles than under a lower calculated level. I just first thought it would be the case, not understanding yet that you throttles are done by powerlevel designs.

It's responsible for current limiting but you should be able to set this in the BIOS usually. The Haswell desktop and mobile had a problem with this setting and EDP unless it was set to 256A or higher which basically disabled current limiting. For Xeon you can similarly I guess, mine is set at 1023. EDP however can cover a lot, not just current throttling. For instance if your VID to multi setting is set above a predetermined value when AVX2 is running. For instance I can have EDP and multi throttling when consuming less than 70W with AVX2 or even faking it to half that value.

If I read out the register I get 0x80000470. This equal to more than 140A powerlimit. Even it hardest work-scenario I barely hit the 110A. So its clear to me I dont hit the Package PPL when I get EDP throttle notification. But also you see that the register is locked. Doesnt matter if I write the register by efi or setting it in UEFI, it wont change (tried it of course).
That EDP includes several kind of throttlings is mentioned and partially described in at least two of the many technical sheets of Haswell. However I would like to know which one actually throttles and I might believe its a per-core based one. Possibly the PL threeshold, but just guessing of course.

It's been over 2 years since I looked at that briefly so I would have to go through it again. My Xeon however doesn't have that MSR so hard for me to be enthusiastic about it. The Southbridge side might have been ME, I'm not 100% though.

Thought you have an E5-v3 ? As you can read in the sdm section E5-v3 0x61E is writable in bit 0 till 3. Sadly whatever bit I write results in a freeze. Even if I write the bit according to my actual setup.
Hoped you could somehow help to get us write the register. Especially as you mentioned in your other thread on notebookreview, but also I mentioned earlier, the PLs clamp with load, the EDP could be pushed further beside increasing single core performance massively.

Did you read this posted earlier?
http://forum.notebookreview.com/threads/tdp-and-power-limiting-haswell.766743/
While it is based on desktop / mobile Haswell the principle is the same.

I just read now, thanks. Another problem of information gathering is, that the information is decentralised and lack crosslinks, but also need to read through hundreds and thousands of post to get a single information you need. Time intensive just.
However, Im still in process of figuring out about mchbar. By myself Im too much bond in private life (work, uni and family all at once)

Sorry I don't understand the question, would you elaborate please.

Ja, I responded to your mentioning having an unlocked multiplier on a Xeon as a reason not looking deeper into 61E-solution for higher clock rates. If I read out the register or just use an software (hwinfo64 for example) to read the information, it clearly responds max-multi is writable on my Xeon, even as it should be disabled by design (stated by Intel). However trying to set it higher results in nothing. So Im not sure about it and was surprised you abandonded the idea.

Well I'm new to Xeon and have never worked in the computer or software industry so not the best person to ask. Of course there's the SDM, while very informative they can have mistakes. There's the CPU data sheets as well and you may find data omitted that was previously disclosed. For instance you may find the answer to your power limiting problem in the lga-2011 datasheet but not in the 2011-3. It's just a hunch. Other areas to look are projects like coreboot. And of course there is always trial and error. For instance, trying to see what happens if a microcode is removed. So far I haven't had anything go too bad that I could not recover from.

Thanks,
some of them I already follow. Coreboot is a bit mysterious still to me. Also found it interesting using clover efi faking cpuid. However you than well know the amount of pages they include. Its hard to work through with limited time.

Dufus, mostly Im interested in increasing somehow the max clockrate, doesnt matter if it happens by an unlocked multi or 61E (the last I would prefer). If we also could push the reason of EDP away to increase allcore multi, would be cool, too. What is your evaluation of succeeding this ? Your knowledge here far more experienced than mine.

 

[GTZ]

I had uploaded,you can try it now.
http://pan.baidu.com/s/1dFkuQel[/QUOTE

Thank you, BIOS 3.6 works perfectly. Now I can fast boot since I no longer need the EFI driver.

 

[kjboughton]

On another Note: does anyone have a code snippet to load uCode or is familiar with the procedure? I would like to add uCode to my drivers to improve OS and Boot compatibility.

http://www.bluestop.org/edk2/docs/trunk/_microcode_update_8c.html

 

[timk1980]

I have Asus Z10PE-D16 WS, and recently upgraded my pair of E5-2666v3 -> E5-2696v3.

I'm noticing a few interesting things: many of the EFI drivers here that do a -50mV don't work... I assume it's just too low voltage for the chip. (I might be able to adjust it in the bios a little higher to overcome that, haven't tried) The only EFI driver anyone has posted that works without issue is v3x2_payne_20_20.efi.

So, my results have been interesting. In particular, with high core counts, the multiplier never goes above a certain point, even if only a few cores are loaded. So, I tried setting to use lower cores in the BIOS, and that definitely helped.

10 cores: 37x on both CPUs
12 cores: 34x and 35x
14 cores: 33x and 34x
16 cores: 32x and 33x
18 cores: 31x and 32x

18 cores is interesting though... it seems to have trouble keeping power up (or thermals maybe? doesn't seem to go much above 70C ever?) and periodically drops from 3000 to 1200 for a few moments on some of the cores, then goes back up and down like a seesaw. The end result is that Cinebench, CPU-Z, Intel Burn test, etc, all do *terribly* by comparison to only having 16 cores enabled, which works amazingly.

I plan to evaluate a better cooling solution soon anyway (Hyper 212 evos right now), but assuming the issue is power related, any thoughts on BIOS settings I can try tweaking, etc? Even in a worst case, the 16 core performance is quite great, and I can switch down to 10 or 12 if I really want higher clocks and am not doing heavily parallel workloads.

Trying to catch up on the many replies on this thread, to see if anyone has seen similar oddities, but figured I'd offer my findings for anyone interested

 

[Kalistoval]

I will prepare some custom bioses in the Next days. Let me know which board exactly and which efi.

Thank you in advance here's a link to my motherboard https://us.msi.com/Motherboard/X99A-GAMING-PRO-CARBON.html. The efi that seems to be working very well is the V3_payne_70_50.efi

 

[Cata40]

After all, what are the differences between C-Payne's efi-es?
To me, what a little, it's about the same, the maximum x34 is on all the core, I mean from -40 to -40 up to -90, just the electricity? Or does the power change? I do not notice any difference

 

[lucien_br]

Note that the C state is turned on，disabled it

I think found out the cause of the crashes. I am not able to remove the microcode in the system32 folder. I am using the Win10 1703 creators version and I am not getting any control over the files of the system32 folder like in win10 1607. Is there any way to Del / Rename file in Creators Edition or is it not necessary to remove the micro code in windows?
Thank you!

 

[lucien_br]

My E5-2696v3 QS works at x34 perfectly fine with these settings (no efi driver), you can try them:

Spoiler: ASUS BIOS settings for E5-2696v3 unlock without efi driver

Thank you so much! It will be very useful for many people. A doubt? Is it necessary to remove the microcode from the system32 folder (mcupdate_genuineintel.dll)? I am not getting permission to del / rename as in win10 1607. I am using Win10 1703 Creators.

 

[moblaw]

to rename mcupdate_genuineintel.dll, you need to boot-safe mode, or change permissions. Either way, I could not boot after changing the name, so I didn't.

I'm using C6 on, EIST on, I'm seeing a 35watt increase. So I've got 23% boost in performance at a cost of 35 watt. That's 11.11 Cinebench R15 mark pr. watt unmodded, and 12,35 Cinebench R15 mark pr. watt modded. I don't dare disable EIST, but does it even correlate to power comsumption? (Guess I should try)

Using the v3x2_50_39_vcc, Works well. Haven't tried anything else.

I've added Paynes' EFI's hope he doesn't mind. Hyperlinks never killed anyone outside Russia.

https://skydrev.org/index.php/s/5rvI5pVwddsXLZT
pw anandtech

 

[Dufus]

If I read out the register I get 0x80000470. This equal to more than 140A powerlimit. Even it hardest work-scenario I barely hit the 110A. So its clear to me I dont hit the Package PPL when I get EDP throttle notification. But also you see that the register is locked. Doesnt matter if I write the register by efi or setting it in UEFI, it wont change (tried it of course).

Which BIOS are you using? There's a bug with some Haswell that can cause EDP throttling even when sufficient current is set so to rule that out we need to set unlimited current.

Thought you have an E5-v3 ? As you can read in the sdm section E5-v3 0x61E is writable in bit 0 till 3. Sadly whatever bit I write results in a freeze. Even if I write the bit according to my actual setup.

Ahh, I assumed you were reading 61E from the CPU. Try the MSR walker under the function tab of this utility http://crystalmark.info/software/CrystalCPUID/index-e.html If you set it to walk from 0x615 to 0x620 do you see 0x61E? If you are trying to write to this MSR and it does not exist a fault will occur in which if not handled may cause a BSOD / freeze.

The total number of Xeon's I have had including this one is one. It would be nice to test on some others, including 306f1 and 306f0 but doesn't look like it is going to happen.

Not MCHBAR for Xeon, 64-bit power limit is also in Bus 0xFF, Dev 1E, Func 0, reg 0xE8

If the Xeon does not support ratio overclocking then setting it will probably be ignored. One can check with HWiNFO under CPU properties. I have only seen this on mobile Haswell and it was mobile CPU's that have been unlocked, not desktop or Xeon that I am aware of. Obviously not many people interested in unlocking mobile CPU's

 

[CANONKONG]

I think found out the cause of the crashes. I am not able to remove the microcode in the system32 folder. I am using the Win10 1703 creators version and I am not getting any control over the files of the system32 folder like in win10 1607. Is there any way to Del / Rename file in Creators Edition or is it not necessary to remove the micro code in windows?
Thank you!

System have microcode will work better, and remove it is not recommended.If the system unstable,close the C state.

 

[lucien_br]

System have microcode will work better, and remove it is not recommended.If the system unstable,close the C state.

Hi. I leave the c3 off and c6 on to run the maximum turbo and it works perfectly with any EFI driver with ucode embedded and even turning off the C-States windows was still crashing with no EFI driver.
I said the original microcode. In Windows 10 Creators update 1703 (crap), I'm not able delete or modify the mcupdate_GenuineIntel.dll. Even if I become the owner of the file, I can not modify it. In the pre-Windows 10 1607 version I was able to modify the mcupdate_GenuineIntel.dll file. Now I loaded ucode 0x39 without deleting mcupdate_GenuineIntel.dll and HWinfo is recognizing "uCU" 39 and I am testing to see if it crashes the windows. So far it seems stable.
Now I only have two doubts:
How do I restore the correct reading of the broken CPU power package with the use of efi driver v3x2_50_39_vcc.efi?
And what is the best microcode to load with cpumcupdate2.1?
Thank you for the help!
Edit:
Windows is still crashing. Any idea how to modify / delete the mcupdate_GenuineIntel.dll file in Windows 10 Creators update 1703?
Edit:
I found a solution with app "Unlocker" and I was able to delete the file delete mcupdate_GenuineIntel.dll and put the ucode 0x39. Testing now.

 

[Cata40]

Hi. I leave the c3 off and c6 on to run the maximum turbo and it works perfectly with any EFI driver with ucode embedded and even turning off the C-States windows was still crashing with no EFI driver.
I said the original microcode. In Windows 10 Creators update 1703 (crap), I'm not able delete or modify the mcupdate_GenuineIntel.dll. Even if I become the owner of the file, I can not modify it. In the pre-Windows 10 1607 version I was able to modify the mcupdate_GenuineIntel.dll file. Now I loaded ucode 0x39 without deleting mcupdate_GenuineIntel.dll and HWinfo is recognizing "uCU" 39 and I am testing to see if it crashes the windows. So far it seems stable.
Now I only have two doubts:
How do I restore the correct reading of the broken CPU power package with the use of efi driver v3x2_50_39_vcc.efi?
And what is the best microcode to load with cpumcupdate2.1?
Thank you for the help!
Edit:
Windows is still crashing. Any idea how to modify / delete the mcupdate_GenuineIntel.dll file in Windows 10 Creators update 1703?
Edit:
I found a solution with app "Unlocker" and I was able to delete the file delete mcupdate_GenuineIntel.dll and put the ucode 0x39. Testing now.

USE UNLOCKER http://www.softpedia.com/get/System/System-Miscellaneous/Unlocker.shtml to modify mcupdate_GenuineIntel.dll, blelieve me best software to delete, modify or rename file can not modify

 

[lucien_br]

USE UNLOCKER http://www.softpedia.com/get/System/System-Miscellaneous/Unlocker.shtml to modify mcupdate_GenuineIntel.dll, blelieve me best software to delete, modify or rename file can not modify

Essential tool for Windows 10 1703 Creators. I try turned owner the file and folder and no way to modify the file or folder System32.
thank you!

 

[plonialmoni]

Hi, I guess I'm a latecomer to this thread, I have an ASRock x99 motherboard. Is all I need to do is flash one of the modded BIOSes for it to get an E5 2690 v3 at max turbo on all cores, or are there more steps I'm missing? I have adequate cooling (240mm water) by the way.